Termux-Vision: On-Device Computer Vision & Multimodal VLM Framework

Native On-Device Computer Vision & Multimodal Vision-Language Model (VLM) Runtime for Android Termux via Direct Bionic libc & Vulkan Compute Acceleration.
Zero PRoot. Zero Virtualization. 100% Native ARMv8.2-A NEON SIMD & Hardware GPU Offloading.

📑 Table of Contents

1. Overview & Key Capabilities
2. Installation Guide & Prebuilt Installer
3. Enabling Hardware GPU Acceleration (with ameva-runtime)
4. Standardized CLI & Parameter Matrix
5. Dual Engine Code Examples (Python & Node.js)
6. Production Diagnostics (termux-vision doctor)
7. Real-World Benchmarks & Hardware Scorecard
8. Memory & VRAM Architecture (Zero CPU-Mapped VRAM)
9. Hardware Requirements & Operational Limits
10. License & Permissible Use

1. Overview & Key Capabilities

termux-vision is an enterprise-grade, on-device multimodal vision inference and spatial computing framework engineered specifically for mobile Android devices. Operating directly against Android's native Bionic libc ABI and host Vulkan compute drivers, termux-vision eliminates heavyweight desktop dependencies (OpenCV, TorchVision) and enables high-throughput visual question answering, OCR image captioning, and classical feature extraction directly on edge hardware.

• 100% Vulkan GPU Compute Canny (0.23 ms): Chains 3-pass SPIR-V compute shaders (Sobel 3x3, NMS, Hysteresis) entirely within VRAM using vkCmdPipelineBarrier, achieving 834x acceleration over Python without CPU memory roundtrips.
• Ultra-Fast ARM64 NEON C++ Kernel (3.02 ms): Permanently eliminates trigonometric atan2f via tangent ratio bit quantization and 1-byte direction buffers, running Canny filtering in 3.02ms on Snapdragon 865 and 4.36ms on Exynos 1380.
• Prebuilt-Asset-First Idempotent Installer (0.005s Skip): Automatically provisions verified precompiled ARM64 native binaries in 2 seconds from official releases, guaranteeing zero-build instant skip if assets already exist.
• 5-Backend Unified CLI Standard: Enforces ['auto', 'gpu', 'vulkan', 'opencl', 'cpu'] and convenience flags (--gpu, --cpu, --opencl) across all subcommands.
• Zero-Deception Fail-Fast Gatekeeper: Strictly rejects defective text-only binaries lacking --mmproj (E015) and corrupted weights (E014), permanently banning silent fallbacks.
• Full-Layer GPU Offloading (-ngl 99): Dispatches all transformer layers and cross-attention vision projections directly to device GPU VRAM (0.00 MiB CPU mapped VRAM).

2. Installation Guide & Prebuilt Installer

termux-vision is distributed across both Python (PyPI) and Node.js (npm) ecosystems, with official precompiled ARM64 wheel assets published on GitHub Releases.

2.1 Termux System Prerequisites

Launch Termux and install required native compilers, Vulkan drivers, and image libraries:

pkg update -y
pkg install -y python nodejs clang make cmake git termux-api wget vulkan-loader vulkan-headers vulkan-tools opencl-headers python-numpy libjpeg-turbo

2.2 Python Package Installation

• Option A: Install from PyPI (Recommended):

  pip install --upgrade pip setuptools wheel
  pip install termux-vision
  

• Option B: Prebuilt Native Engine Provisioning (Idempotent 0.005s):

  # Automatically download & unpack verified ARM64 prebuilt assets
  termux-vision install
  
  # Optional maintenance flags:
  # termux-vision install --force        # Force re-downloading prebuilts
  # termux-vision install --from-source   # Force compiling from local C++ source
  # termux-vision install --dry-run       # Check integrity without making changes
  

• Option C: Direct GitHub Releases Wheel Asset:

  # Download and install the prebuilt v1.5.0 release wheel
  pip install https://github.com/uno-km/termux-vision/releases/download/v1.5.0/termux_vision-1.5.0-py3-none-any.whl
  

2.3 Node.js / TypeScript CLI Installation

# Global CLI installation
npm install -g termux-vision

# Local project dependency
npm install termux-vision

2.4 One-Line Bootstrap Installer

Run the universal bootstrap installer to automatically configure repositories, compile native C/C++ acceleration shims, and verify hardware:

curl -sL https://raw.githubusercontent.com/uno-km/termux-vision/main/install.sh | bash

3. Enabling Hardware GPU Acceleration (with ameva-runtime)

To unlock mobile GPU acceleration via Vulkan compute shaders and achieve significant speedups over pure CPU execution, install termux-vision alongside ameva-runtime:

🌟 One-Line Installation

# Python Environment
pip install termux-vision ameva-runtime termux-llamacpp

# Node.js Environment
npm install -g termux-vision @ameva/runtime

🔮 Mobile GPU Silicon Architecture Status

Verify GPU driver detection and hardware readiness:

termux-vision doctor

4. Standardized CLI & Parameter Matrix

termux-vision strictly complies with the official uno-km family 5-backend CLI standard:

Practical CLI Usage Examples

# 1. 100% Vulkan GPU Canny Edge Detection (0.23 ms on Adreno 830)
termux-vision canny photo.jpg -o edges.png --gpu --low 40 --high 120

# 2. Ultra-Fast NEON C++ Canny Edge Detection (3.02 ms on S20 CPU)
termux-vision canny photo.jpg -o edges.png --cpu

# 3. Multimodal VLM Inference with Automated GPU Routing
termux-vision vlm photo.jpg -p "What objects are visible in this scene?"

# 4. Pure GPU Mode on ARM Mali Silicon (Galaxy S21 / A35)
termux-vision vlm photo.jpg -d gpu --tune-mali -p "Describe the text and layout."

# 5. Prebuilt Native Binary Provisioning (0.005s Idempotent Skip)
termux-vision install

5. Dual Engine Code Examples (Python & Node.js)

5.1 Python SDK

import termux_vision as tv

# 1. Hardware-Accelerated Canny Edge Detection (0.23ms Vulkan GPU / 3.02ms NEON CPU)
image = tv.io.load_image("document.jpg")
grayscale = tv.transforms.to_grayscale(image)
edges = tv.cv.canny(grayscale, low_threshold=40, high_threshold=120, backend="auto")
tv.io.save_image(edges, "edges.png")

# 2. On-Device Multimodal VLM Inference (Vulkan GPU Accelerated)
with tv.vlm.load("smolvlm-500m", device="gpu") as engine:
    result = engine.describe(
        "document.jpg",
        prompt="Extract all visible text and summarize key bullet points.",
        quality="optimal",
        max_tokens=200
    )
    print(f"Backend: {result.metrics.backend} | TPS: {result.metrics.tokens_per_second:.2f} tok/s")
    print(f"Response:\n{result.text}")

5.2 Node.js / TypeScript SDK

import tv from 'termux-vision';

// 1. Hardware Diagnostic Probe
const doctor = tv.doctor(true);
console.log(`Vulkan GPU: ${doctor.vulkan.status} | Available RAM: ${doctor.hardware.availableRamMb} MB`);

// 2. Multimodal VLM Inference
const engine = await tv.load({ modelId: 'smolvlm-500m', device: 'gpu' });
const response = await engine.describe('photo.jpg', {
  prompt: 'Identify the geometric shapes and colors.',
  quality: 'optimal',
  maxTokens: 100
});

console.log(`[${response.metrics.backend.toUpperCase()}] ${response.text}`);
engine.close();

6. Production Diagnostics (termux-vision doctor)

Termux-Vision features an integrated hardware diagnostics probe to inspect the host environment before launching inference:

termux-vision doctor

Diagnostic Output Profile:

=== termux-vision Diagnostic Doctor ===
  Platform : Linux (aarch64) | Android: True
  RAM      : Total 7812MB | Available 3450MB
  CPU Cores: 8 (big.LITTLE Affinity Governor active)
  Vulkan   : Loader=True | Driver=/system/lib64/libvulkan.so | Status=READY
  GPU Soc  : ARM Mali-G78 MP14 (Exynos 2100)
  Models   : 2 installed in ~/.cache/termux-vision/models
  Preset   : optimal (768px recommended)

7. Real-World Benchmarks & Hardware Scorecard

7.1 Classical Vision Filtering Latency (512x512 Image, Physical Devices)

7.2 On-Device Multimodal VLM Benchmark

8. Memory & VRAM Architecture

+---------------------------------------------------------------+
|             Physical Mobile LPDDR4X/LPDDR5 RAM (8 GB)          |
+---------------------------------------------------------------+
   |                                                    |
   v                                                    v
+-------------------------------+       +-------------------------------+
|      Android OS & Framework   |       |       Termux User Space       |
|          (~3.5 - 4.2 GB)      |       |          (~3.8 - 4.5 GB)      |
+-------------------------------+       +-------------------------------+
                                                        |
                                                        v
                                        +-------------------------------+
                                        |    Vulkan Unified Memory      |
                                        |  - Model Weights: 1059.02 MiB |
                                        |  - KV Cache     :  384.00 MiB |
                                        |  - CPU Mapped   :    0.00 MiB |
                                        +-------------------------------+

9. Hardware Requirements & Operational Limits

10. License & Permissible Use

Licensed under the Apache License, Version 2.0.
Copyright (c) 2026 Eunho Kim (@uno-km) & AMEVA Open-Source Foundation (AOSF).

• Official Documentation Portal
• AMEVA Open-Source Foundation
• GitHub Repository
• Issue Tracker